Refine
Document Type
- Article (17)
Language
- English (17)
Has Fulltext
- yes (17)
Is part of the Bibliography
- no (17)
Keywords
- Artificial intelligence (2)
- Brain metastasis (2)
- CT (2)
- Radiomics (2)
- 4-1BB (1)
- Anemia (1)
- Angiography (1)
- BBB (1)
- BPH (1)
- BPO (1)
Institute
- Medizin (17) (remove)
Simple Summary: Early and accurate diagnosis of breast cancer that has spread to other organs and tissues is crucial, as therapeutic decisions and outcome expectations might change. Computed tomography (CT) is often used to detect breast cancer’s spread, but this method has its weaknesses. The computer-assisted technique “radiomics” extracts grey-level patterns, so-called radiomic features, from medical images, which may reflect underlying biological processes. Our retrospective study therefore evaluated whether breast cancer spread can be predicted by radiomic features derived from iodine maps, an application on a new generation of CT scanners visualizing tissue blood flow. Based on 77 patients with newly diagnosed breast cancer, we found that this approach might indeed predict cancer spread to other organs/tissues. In the future, radiomics may serve as an additional tool for cancer detection and risk assessment.
Abstract: Dual-energy CT (DECT) iodine maps enable quantification of iodine concentrations as a marker for tissue vascularization. We investigated whether iodine map radiomic features derived from staging DECT enable prediction of breast cancer metastatic status, and whether textural differ- ences exist between primary breast cancers and metastases. Seventy-seven treatment-naïve patients with biopsy-proven breast cancers were included retrospectively (41 non-metastatic, 36 metastatic). Radiomic features including first-, second-, and higher-order metrics as well as shape descriptors were extracted from volumes of interest on iodine maps. Following principal component analysis, a multilayer perceptron artificial neural network (MLP-NN) was used for classification (70% of cases for training, 30% validation). Histopathology served as reference standard. MLP-NN predicted metastatic status with AUCs of up to 0.94, and accuracies of up to 92.6 in the training and 82.6 in the validation datasets. The separation of primary tumor and metastatic tissue yielded AUCs of up to 0.87, with accuracies of up to 82.8 in the training, and 85.7 in the validation dataset. DECT iodine map-based radiomic signatures may therefore predict metastatic status in breast cancer patients. In addition, microstructural differences between primary and metastatic breast cancer tissue may be reflected by differences in DECT radiomic features.
Background: To assess the potential of radiomic features to quantify components of blood in intraaortic vessels to non-invasively predict moderate-to-severe anemia in non-contrast enhanced CT scans. Methods: One hundred patients (median age, 69 years; range, 19–94 years) who received CT scans of the thoracolumbar spine and blood-testing for hemoglobin and hematocrit levels ± 24 h between 08/2018 and 11/2019 were retrospectively included. Intraaortic blood was segmented using a spherical volume of interest of 1 cm diameter with consecutive radiomic analysis applying PyRadiomics software. Feature selection was performed applying analysis of correlation and collinearity. The final feature set was obtained to differentiate moderate-to-severe anemia. Random forest machine learning was applied and predictive performance was assessed. A decision-tree was obtained to propose a cut-off value of CT Hounsfield units (HU). Results: High correlation with hemoglobin and hematocrit levels was shown for first-order radiomic features (p < 0.001 to p = 0.032). The top 3 features showed high correlation to hemoglobin values (p) and minimal collinearity (r) to the top ranked feature Median (p < 0.001), Energy (p = 0.002, r = 0.387), Minimum (p = 0.032, r = 0.437). Median (p < 0.001) and Minimum (p = 0.003) differed in moderate-to-severe anemia compared to non-anemic state. Median yielded superiority to the combination of Median and Minimum (p(AUC) = 0.015, p(precision) = 0.017, p(accuracy) = 0.612) in the predictive performance employing random forest analysis. A Median HU value ≤ 36.5 indicated moderate-to-severe anemia (accuracy = 0.90, precision = 0.80). Conclusions: First-order radiomic features correlate with hemoglobin levels and may be feasible for the prediction of moderate-to-severe anemia. High dimensional radiomic features did not aid augmenting the data in our exemplary use case of intraluminal blood component assessment.
Purpose: To stratify differences in visual semantic and quantitative imaging features in intensive care patients with nonspecific mastoid effusions versus patients with acute mastoiditis (AM) requiring surgical treatment. Methods: We included 48 patients (male, 28; female, 20; mean age, 59.5 ± 18.1 years) with mastoid opacification (AM, n = 24; control, n = 24) who underwent clinically indicated cerebral CT between 12/2007 and 07/2018 in this retrospective study. Semantic features described the extend and asymmetry of mastoid and middle-ear cavity opacification and complications like erosive changes. Minimum, maximum and mean Hounsfield unit (HU) values were obtained as quantitative features. We analyzed the features employing univariate testing. Results: Compared to intensive care patients, AM patients revealed asymmetric mastoid or middle-ear cavity opacification (likelihood-ratio (LR) < 0.001). Applying a dedicated threshold of the extent of opacification, AM patients reached significance levels of LR = 0.042 and 0.002 for mastoid and middle-ear cavity opacification. AM cases showed higher maximum and mean HU values (p = 0.009, p = 0.024). Conclusions: We revealed that the extent and asymmetry of mastoid and middle-ear cavity opacification differs significantly between AM patients and intensive care patients. Multicenter research is needed to expand our cohort and possibly pave the way to build a non-invasive predictive model for AM in the future.
Purpose: To identify transjugular intrahepatic portosystemic shunt (TIPS) thrombosis in abdominal CT scans applying quantitative image analysis.
Materials and methods: We retrospectively screened 184 patients to include 20 patients (male, 8; female, 12; mean age, 60.7 ± 8.87 years) with (case, n = 10) and without (control, n = 10) in-TIPS thrombosis who underwent clinically indicated contrast-enhanced and unenhanced abdominal CT followed by conventional TIPS-angiography between 08/2014 and 06/2020. First, images were scored visually. Second, region of interest (ROI) based quantitative measurements of CT attenuation were performed in the inferior vena cava (IVC), portal vein and in four TIPS locations. Minimum, maximum and average Hounsfield unit (HU) values were used as absolute and relative quantitative features. We analyzed the features with univariate testing.
Results: Subjective scores identified in-TIPS thrombosis in contrast-enhanced scans with an accuracy of 0.667 – 0.833. Patients with in-TIPS thrombosis had significantly lower average (p < 0.001), minimum (p < 0.001) and maximum HU (p = 0.043) in contrast-enhanced images. The in-TIPS / IVC ratio in contrast-enhanced images was significantly lower in patients with in-TIPS thrombosis (p < 0.001). No significant differences were found for unenhanced images. Analyzing the visually most suspicious ROI with consecutive calculation of its ratio to the IVC, all patients with a ratio < 1 suffered from in-TIPS thrombosis (p < 0.001, sensitivity and specificity = 100%).
Conclusion: Quantitative analysis of abdominal CT scans facilitates the stratification of in-TIPS thrombosis. In contrast-enhanced scans, an in-TIPS / IVC ratio < 1 could non-invasively stratify all patients with in-TIPS thrombosis.
Highlights
• MRI and ultrasound provided significant correlations between findings suggestive of vasculitis and the final diagnosis.
• Careful selection of available imaging techniques is warranted considering the time course, location, and clinical history.
• Considering its moderate diagnostic power to distinguish tracer uptake, a holistic view of PET/CT findings is essential.
Abstract
Purpose: To assess the diagnostic value of different imaging modalities in distinguishing systemic vasculitis from other internal and immunological diseases.
Methods: This retrospective study included 134 patients with suspected vasculitis who underwent ultrasound, magnetic resonance imaging (MRI), or 18F-fluorodeoxyglucose positron emission tomography/computed tomography (18F-FDG PET/CT) between 01/2010 and 01/2019, finally consisting of 70 individuals with vasculitis. The main study parameter was the confirmation of the diagnosis using one of the three different imaging modalities, with the adjudicated clinical and histopathological diagnosis as the gold standard. A secondary parameter was the morphological appearance of the vessel affected by vasculitis.
Results: Patients with systemic vasculitis had myriad clinical manifestations with joint pain as the most common symptom. We found significant correlations between different imaging findings suggestive of vasculitis and the final adjudicated clinical diagnosis. In this context, on MRI, vessel wall thickening, edema, and diameter differed significantly between vasculitis and non-vasculitis groups (p < 0.05). Ultrasound revealed different findings that may serve as red flags in identifying patients with vasculitis, such as vascular occlusion or halo sign (p = 0.02 vs. non-vasculitis group). Interestingly, comparing maximal standardized uptake values from PET/CT examinations with vessel wall thickening or vessel diameter did not result in significant differences (p > 0.05).
Conclusions: We observed significant correlations between different imaging findings suggestive of vasculitis on ultrasound or MRI and the final adjudicated diagnosis. While ultrasound and MRI were considered suitable imaging methods for detecting and discriminating typical vascular changes, 18F-FDG PET/CT requires careful timing and patient selection given its moderate diagnostic accuracy.
Objective: To investigate the value of standard [digital rectal examination (DRE), PSA] and advanced (mpMRI, prostate biopsy) clinical evaluation for prostate cancer (PCa) detection in contemporary patients with clinical bladder outlet obstruction (BOO) scheduled for Holmium laser enucleation of the prostate (HoLEP).
Material and Methods: We retrospectively analyzed 397 patients, who were referred to our tertiary care laser center for HoLEP due to BOO between 11/2017 and 07/2020. Of those, 83 (20.7%) underwent further advanced clinical PCa evaluation with mpMRI and/or prostate biopsy due to elevated PSA and/or lowered PSA ratio and/or suspicious DRE. Logistic regression and binary regression tree models were applied to identify PCa in BOO patients.
Results: An mpMRI was conducted in 56 (66%) of 83 patients and revealed PIRADS 4/5 lesions in 14 (25%) patients. Subsequently, a combined systematic randomized and MRI-fusion biopsy was performed in 19 (23%) patients and revealed in PCa detection in four patients (5%). A randomized prostate biopsy was performed in 31 (37%) patients and revealed in PCa detection in three patients (4%). All seven patients (9%) with PCa detection underwent radical prostatectomy with 29% exhibiting non-organ confined disease. Incidental PCa after HoLEP (n = 76) was found in nine patients (12%) with advanced clinical PCa evaluation preoperatively. In univariable logistic regression analyses, PSA, fPSA ratio, and PSA density failed to identify patients with PCa detection. Conversely, patients with a lower International Prostate Symptom Score (IPSS) and PIRADs 4/5 lesion in mpMRI were at higher risk for PCa detection. In multivariable adjusted analyses, PIRADS 4/5 lesions were confirmed as an independent risk factor (OR 9.91, p = 0.04), while IPSS did not reach significance (p = 0.052).
Conclusion: In advanced clinical PCa evaluation mpMRI should be considered in patients with elevated total PSA or low fPSA ratio scheduled for BOO treatment with HoLEP. Patients with low IPSS or PIRADS 4/5 lesions in mpMRI are at highest risk for PCa detection. In patients with a history of two or more sets of negative prostate biopsies, advanced clinical PCa evaluation might be omitted.
Higher grade meningiomas tend to recur. We aimed to evaluate protein levels of vascular endothelial growth factor (VEGF)-A with the VEGF-receptors 1-3 and the co-receptors Neuropilin (NRP)-1 and -2 in WHO grade II and III meningiomas to elucidate the rationale for targeted treatments. We investigated 232 specimens of 147 patients suffering from cranial meningioma, including recurrent tumors. Immunohistochemistry for VEGF-A, VEGFR-1-3, and NRP-1/-2 was performed on tissue micro arrays. We applied a semiquantitative score (staining intensity x frequency). VEGF-A, VEGFR-1-3, and NRP-1 were heterogeneously expressed. NRP-2 was mainly absent. We demonstrated a significant increase of VEGF-A levels on tumor cells in WHO grade III meningiomas (p = 0.0098). We found a positive correlation between expression levels of VEGF-A and VEGFR-1 on tumor cells and vessels (p < 0.0001). In addition, there was a positive correlation of VEGF-A and VEGFR-3 expression on tumor vessels (p = 0.0034). VEGFR-2 expression was positively associated with progression-free survival (p = 0.0340). VEGF-A on tumor cells was negatively correlated with overall survival (p = 0.0084). The VEGF-A-driven system of tumor angiogenesis might still present a suitable target for adjuvant therapy in malignant meningioma disease. However, its role in malignant tumor progression may not be as crucial as expected. The value of comprehensive testing of the ligand and all receptors prior to administration of anti-angiogenic therapy needs to be evaluated in clinical trials.
Purpose: To test the effect of anatomic variants of the prostatic apex overlapping the membranous urethra (Lee type classification), as well as median urethral sphincter length (USL) in preoperative multiparametric magnetic resonance imaging (mpMRI) on the very early continence in open (ORP) and robotic-assisted radical prostatectomy (RARP) patients. Methods: In 128 consecutive patients (01/2018–12/2019), USL and the prostatic apex classified according to Lee types A–D in mpMRI prior to ORP or RARP were retrospectively analyzed. Uni- and multivariable logistic regression models were used to identify anatomic characteristics for very early continence rates, defined as urine loss of ≤ 1 g in the PAD-test. Results: Of 128 patients with mpMRI prior to surgery, 76 (59.4%) underwent RARP vs. 52 (40.6%) ORP. In total, median USL was 15, 15 and 10 mm in the sagittal, coronal and axial dimensions. After stratification according to very early continence in the PAD-test (≤ 1 g vs. > 1 g), continent patients had significantly more frequently Lee type D (71.4 vs. 54.4%) and C (14.3 vs. 7.6%, p = 0.03). In multivariable logistic regression models, the sagittal median USL (odds ratio [OR] 1.03) and Lee type C (OR: 7.0) and D (OR: 4.9) were independent predictors for achieving very early continence in the PAD-test. Conclusion: Patients’ individual anatomical characteristics in mpMRI prior to radical prostatectomy can be used to predict very early continence. Lee type C and D suggest being the most favorable anatomical characteristics. Moreover, longer sagittal median USL in mpMRI seems to improve very early continence rates.
Our purpose was to analyze the robustness and reproducibility of magnetic resonance imaging (MRI) radiomic features. We constructed a multi-object fruit phantom to perform MRI acquisition as scan-rescan using a 3 Tesla MRI scanner. We applied T2-weighted (T2w) half-Fourier acquisition single-shot turbo spin-echo (HASTE), T2w turbo spin-echo (TSE), T2w fluid-attenuated inversion recovery (FLAIR), T2 map and T1-weighted (T1w) TSE. Images were resampled to isotropic voxels. Fruits were segmented. The workflow was repeated by a second reader and the first reader after a pause of one month. We applied PyRadiomics to extract 107 radiomic features per fruit and sequence from seven feature classes. We calculated concordance correlation coefficients (CCC) and dynamic range (DR) to obtain measurements of feature robustness. Intraclass correlation coefficient (ICC) was calculated to assess intra- and inter-observer reproducibility. We calculated Gini scores to test the pairwise discriminative power specific for the features and MRI sequences. We depict Bland Altmann plots of features with top discriminative power (Mann–Whitney U test). Shape features were the most robust feature class. T2 map was the most robust imaging technique (robust features (rf), n = 84). HASTE sequence led to the least amount of rf (n = 20). Intra-observer ICC was excellent (≥ 0.75) for nearly all features (max–min; 99.1–97.2%). Deterioration of ICC values was seen in the inter-observer analyses (max–min; 88.7–81.1%). Complete robustness across all sequences was found for 8 features. Shape features and T2 map yielded the highest pairwise discriminative performance. Radiomics validity depends on the MRI sequence and feature class. T2 map seems to be the most promising imaging technique with the highest feature robustness, high intra-/inter-observer reproducibility and most promising discriminative power.
Objectives: To analyze the performance of radiological assessment categories and quantitative computational analysis of apparent diffusion coefficient (ADC) maps using variant machine learning algorithms to differentiate clinically significant versus insignificant prostate cancer (PCa). Methods: Retrospectively, 73 patients were included in the study. The patients (mean age, 66.3 ± 7.6 years) were examined with multiparametric MRI (mpMRI) prior to radical prostatectomy (n = 33) or targeted biopsy (n = 40). The index lesion was annotated in MRI ADC and the equivalent histologic slides according to the highest Gleason Grade Group (GrG). Volumes of interest (VOIs) were determined for each lesion and normal-appearing peripheral zone. VOIs were processed by radiomic analysis. For the classification of lesions according to their clinical significance (GrG ≥ 3), principal component (PC) analysis, univariate analysis (UA) with consecutive support vector machines, neural networks, and random forest analysis were performed. Results: PC analysis discriminated between benign and malignant prostate tissue. PC evaluation yielded no stratification of PCa lesions according to their clinical significance, but UA revealed differences in clinical assessment categories and radiomic features. We trained three classification models with fifteen feature subsets. We identified a subset of shape features which improved the diagnostic accuracy of the clinical assessment categories (maximum increase in diagnostic accuracy ΔAUC = + 0.05, p < 0.001) while also identifying combinations of features and models which reduced overall accuracy. Conclusions: The impact of radiomic features to differentiate PCa lesions according to their clinical significance remains controversial. It depends on feature selection and the employed machine learning algorithms. It can result in improvement or reduction of diagnostic performance.